Handy AI is a reader-supported publication. To receive new posts and support our work, consider becoming a free or paid subscriber.
what to know for now
🚀 OpenAI preparing to launch GPT-4.1 and new reasoning models. OpenAI is reportedly getting ready to unveil several new AI models this week, including GPT-4.1, a revamped version of their GPT-4o multimodal model, alongside smaller GPT-4.1 mini and nano versions. The company is also preparing to release the full version of its o3 reasoning model and an o4 mini version, with references to these models already appearing in the ChatGPT web interface. Read more
🧠 ChatGPT gets comprehensive memory upgrade. OpenAI has rolled out a significant update to ChatGPT that allows the AI to remember and reference all past conversations across separate chat sessions. This new memory system goes beyond the previous manual memory feature, automatically creating connections between historical interactions spanning months or potentially years. Read more
🎬 Google Cloud expands generative AI media tools. Google Cloud has announced significant updates to Vertex AI, adding comprehensive generative media capabilities across video, image, speech, and music. The updates include Lyria for text-to-music generation, Veo 2 with enhanced video editing controls, Chirp 3 with Instant Custom Voice creation, and Imagen 3 with improved image generation and inpainting features. Read more
💰 OpenAI reaches 800 million users. OpenAI CEO Sam Altman has announced that the company now has approximately 800 million users across its AI products, with ChatGPT accounting for the majority of this user base. The rapid growth represents a significant milestone for the company, which launched ChatGPT less than two years ago and has seen its valuation soar to $150 billion. Read more
Jake's Take: These researchers have developed OmniSVG, a novel diffusion-based approach for generating high-quality Scalable Vector Graphics (SVG) directly from text prompts or raster images. The method uses a two-stage process: first generating a rasterized preview with a diffusion model, then converting it to SVG using a specialized decoder that produces clean, editable vector graphics with precise paths and shapes. The system handles complex visual elements including gradients, patterns, and typography while maintaining editability.
SVGs are a vector format that are used in web design and development, and have notoriously been difficult for AI models to generate. Unlike previous approaches that produce raster images requiring manual vectorization, OmniSVG creates production-ready vector graphics that designers can immediately incorporate into their projects, helping to shorten the gap between AI image gen and professional design workflows.
what to know for later
🚀 Sutskever's Safe Superintelligence valued at $32B. OpenAI co-founder Ilya Sutskever's startup Safe Superintelligence (SSI) has reportedly reached a valuation of $32 billion in its latest funding round, despite being less than a year old. The company, focused on developing AI systems with robust safety guarantees, has attracted significant investment from major tech VCs and sovereign wealth funds. This valuation places SSI among the fastest companies to reach unicorn status in history. Read more
🤖 OpenAI developing "A-SWE" agentic software engineer. OpenAI CFO Sarah Friar has confirmed the company is working on its third AI agent called "A-SWE" (Agentic Software Engineer), designed to build applications, perform quality assurance, conduct bug testing, and create documentation. This AI agent aims to go beyond augmenting human engineers to potentially handling complete development workflows independently. Read more
💰 Murati reportedly seeking massive AI startup funding. Former OpenAI CTO Mira Murati is rumored to be in talks to raise up to $2 billion for her new AI venture, which would set a record for the largest seed funding in history if completed. The startup, which remains in stealth mode, is reportedly developing novel AI architecture that addresses fundamental limitations in current large language models. Read more
🔄 OpenAI considers acquiring Altman-Ive hardware startup. OpenAI is reportedly exploring the acquisition of the AI hardware startup co-founded by CEO Sam Altman and former Apple design chief Jony Ive. The startup is developing consumer AI devices intended to offer a more natural interface for AI interactions. This potential vertical integration could give OpenAI greater control over the full AI experience from software to hardware. Read more
🏭 Nvidia to manufacture AI supercomputers in US. Nvidia has announced plans to produce its AI supercomputers entirely in the United States, commissioning over a million square feet of manufacturing space in Arizona and Texas. The company has started production of Blackwell chips at TSMC's Arizona plants and is building supercomputer manufacturing facilities with Foxconn in Houston and Wistron in Dallas, with mass production expected within 12-15 months. Read more
